Custom Policy-Based Authorization

Underneath the covers the role authorization and claims authorization make use of a requirement, a handler for the requirement and a pre-configured policy. These building blocks allow you to express authorization evaluations in code, allowing for a richer, reusable, and easily testable authorization structure.

An authorization policy is made up of one or more requirements and registered at application startup as part of the Authorization service configuration, in ConfigureServices in the Startup.cs file.

public void ConfigureServices(IServiceCollection services)
{
    services.AddMvc();

    services.AddAuthorization(options =>
    {
        options.AddPolicy("Over21",
                          policy => policy.Requirements.Add(new MinimumAgeRequirement(21)));
    }
});

Here you can see an “Over21” policy is created with a single requirement, that of a minimum age, which is passed as a parameter to the requirement.

Policies are applied using the Authorize attribute by specifying the policy name, for example;

[Authorize(Policy="Over21")]
public class AlcoholPurchaseRequirementsController : Controller
{
    public ActionResult Login()
    {
    }

    public ActionResult Logout()
    {
    }
}

Requirements

An authorization requirement is a collection of data parameters that a policy can use to evaluate the current user principal. In our Minimum Age policy the requirement we have a single parameter, the minimum age. A requirement must implement IAuthorizationRequirement. This is an empty, marker interface. A parameterized minimum age requirement might be implemented as follows;

public class MinimumAgeRequirement : IAuthorizationRequirement
{
    public MinimumAgeRequirement(int age)
    {
        MinimumAge = age;
    }

    protected int MinimumAge { get; set; }
}

A requirement doesn’t need to have data or properties.

Authorization Handlers

An authorization handler is responsible for the evaluation of any properties of a requirement. The authorization handler must evaluate them against a provided AuthorizationContext to decide if authorization is allowed. A requirement can have multiple handlers. Handlers must inherit AuthorizationHandler<T> where T is the requirement it handles.

The minimum age handler might look like this:

public class MinimumAgeHandler : AuthorizationHandler<MinimumAgeRequirement>
{
    protected override Task HandleRequirementAsync(AuthorizationContext context, MinimumAgeRequirement requirement)
    {
        if (!context.User.HasClaim(c => c.Type == ClaimTypes.DateOfBirth &&
                                   c.Issuer == "http://contoso.com"))
        {
            return Task.FromResult(0);
        }

        var dateOfBirth = Convert.ToDateTime(context.User.FindFirst(
            c => c.Type == ClaimTypes.DateOfBirth && c.Issuer == "http://contoso.com").Value);

        int calculatedAge = DateTime.Today.Year - dateOfBirth.Year;
        if (dateOfBirth > DateTime.Today.AddYears(-calculatedAge))
        {
            calculatedAge--;
        }

        if (calculatedAge >= requirement.MinimumAge)
        {
            context.Succeed(requirement);
        }
        return Task.FromResult(0);
    }
}

In the code above we first look to see if the current user principal has a date of birth claim which has been issued by an Issuer we know and trust. If the claim is missing we can’t authorize so we return. If we have a claim, we figure out how old the user is, and if they meet the minimum age passed in by the requirement then authorization has been successful. Once authorization is successful we call context.Succeed() passing in the requirement that has been successful as a parameter.

Handlers must be registered in the services collection during configuration, for example;

public void ConfigureServices(IServiceCollection services)
{
    services.AddMvc();

    services.AddAuthorization(options =>
    {
        options.AddPolicy("Over21",
                          policy => policy.Requirements.Add(new MinimumAgeRequirement(21)));
    });

    services.AddSingleton<IAuthorizationHandler, MinimumAgeHandler>();
}

Each handler is added to the services collection by using services.AddSingleton<IAuthorizationHandler, YourHandlerClass>(); passing in your handler class.

What should a handler return?

You can see in our handler example that the Handle() method has no return value, so how do we indicate success or failure?

  • A handler indicates success by calling context.Succeed(IAuthorizationRequirement requirement), passing the requirement that has been successfully validated.
  • A handler does not need to handle failures generally, as other handlers for the same requirement may succeed.
  • To guarantee failure even if other handlers for a requirement succeed, call context.Fail.

Regardless of what you call inside your handler all handlers for a requirement will be called when a policy requires the requirement. This allows requirements to have side effects, such as logging, which will always take place even if context.Fail() has been called in another handler.

Why would I want multiple handlers for a requirement?

In cases where you want evaluation to be on an OR basis you implement multiple handlers for a single requirement. For example, Microsoft has doors which only open with key cards. If you leave your key card at home the receptionist prints a temporary sticker and opens the door for you. In this scenario you’d have a single requirement, EnterBuilding, but multiple handlers, each one examining a single requirement.

public class EnterBuildingRequirement : IAuthorizationRequirement
{
}

public class BadgeEntryHandler : AuthorizationHandler<EnterBuildingRequirement>
{
    protected override Task HandleRequirementAsync(AuthorizationContext context, EnterBuildingRequirement requirement)
    {
        if (context.User.HasClaim(c => c.Type == ClaimTypes.BadgeId &&
                                       c.Issuer == "http://microsoftsecurity"))
        {
            context.Succeed(requirement);
            return Task.FromResult(0);
        }
    }
}

public class HasTemporaryStickerHandler : AuthorizationHandler<EnterBuildingRequirement>
{
    protected override Task HandleRequirementAsync(AuthorizationContext context, EnterBuildingRequirement requirement)
    {
        if (context.User.HasClaim(c => c.Type == ClaimTypes.TemporaryBadgeId &&
                                       c.Issuer == "https://microsoftsecurity"))
        {
            // We'd also check the expiration date on the sticker.
            context.Succeed(requirement);
            return Task.FromResult(0);
        }
    }
}

Now, assuming both handlers are registered when a policy evaluates the EnterBuildingRequirement if either handler succeeds the policy evaluation will succeed.

Using a func to fufill a policy

There may be occasions where fufilling a policy is simple to express in code. It is possible to simply supply a Func<AuthorizationHandlerContext, bool> when configuring your policy with the RequireAssertion policy builder.

For example the previous BadgeEntryHandler could be rewritten as follows;

services.AddAuthorization(options =>
    {
        options.AddPolicy("BadgeEntry",
                          policy => policy.RequireAssertion(context =>
                                  context.User.HasClaim(c =>
                                     (c.Type == ClaimTypes.BadgeId ||
                                      c.Type == ClaimTypes.TemporaryBadgeId)
                                      && c.Issuer == "https://microsoftsecurity"));
                          }));
    }
 }

Accessing MVC Request Context In Handlers

The Handle method you must implement in an authorization handler has two parameters, an AuthorizationContext and the Requirement you are handling. Frameworks such as MVC or Jabbr are free to add any object to the Resource property on the AuthorizationContext to pass through extra information.

For example MVC passes an instance of Microsoft.AspNetCore.Mvc.Filters.AuthorizationFilterContext in the resource property which is used to access HttpContext, RouteData and everything else MVC provides.

The use of the Resource property is framework specific. Using information in the Resource property will limit your authorization policies to particular frameworks. You should cast the Resource property using the as keyword, and then check the cast has succeed to ensure your code doesn’t crash with InvalidCastExceptions when run on other frameworks;

var mvcContext = context.Resource as Microsoft.AspNetCore.Mvc.Filters.AuthorizationFilterContext;

if (mvcContext != null)
{
    // Examine MVC specific things like routing data.
}